Information and Referral servces for the elderly have proliferated in recent years. Preliminary evidence suggests that they have been ineffectual in reaching and serving homebound elderly and others suffering from multiple problems. Two operating I & R Programs for the elderly are compared for their relative effectivenss. One program uses telephone counselors (both professional and non-professional), the other relies on face-to-face contact with age and racial peers. Examined are the relative effectiveness of these systems in assessing needs, making adequate referrals, and in locating the dysfunctional population. The productivity of these systems is compared in terms of per unit cost, and client determined service satisfaction. In-take questionnaires detailing age, sex, race, and service problems, are administered to approximately 1,000 I & R system users monthly. A purposeful sample of clients will be drawn from this universe. Forced-order stepwise multiple regression is used to test the effect on service outcomes. These outcomes include, type of referrals made, service provided, appropriateness of service, cost/length of contact needed to realize outcome.